Today I learned that the domain name in an ad can have a MASSIVE affect on CTR. I assumed it could make a difference but I've never tested it before, and never realized it could make such a big difference... I launched a new site, and I basically copied an existing adwords campaign for a similar site over to the new adwords account for the new site. I was in a hurry setting up my ads and long story short I changed just the display URL in my ad, and left and the rest of the ad the same. After running for a few days on the SAME keywords and roughly the same average position, with the SAME ad with only the display URL different, the new ad literally got a CTR that is HALF of the ad for my other site. CTR cut in half just because of the display URL? Dayum! So now I have one more thing to test. Great. So what is the best way to test different domain names in display URLs without having to setup actual sites on each of the domains? I can't really do this because the site is transaction based, and not just a few pages of HTML, etc. For testing purposes can I get away with just registering a bunch of domains and forwarding or redirecting them all to my main site? I just need to see which one helps my ad CTR the best, and then I will make that my official domain for the site. Even more evil, could I get away with just making up display urls and keeping the destination url the same for a few days? (i.e. the display URL and destination URLs are not even the same domain) Or will G catch this immediately? I'm assuming Google doesn't encourage this but at the same time I assume I could get away with it for at least a few days which would give me enough data for my test. What about my quality score as it pertains to my landing page, if I do a redirect? Will G base my quality score on the final URL that the user gets redirected to, or will my quality score be horrible since my destination URL is just a redirect? Is there a better way to do this? Ideally one that is Google-approved? Thanks!
You also have to pay attention to average position - By creating a new campaign, you lose the history, and I wouldn't be surprised if your position was lower on the new ad. As for how to test most effectively, just attach another domain (perhaps called a "parked" domain) to your hosting account, so all pages on olddomain.com are the same as on newdomain.com.
Did you do a true A-B split-test on the different display URLs? To get accurate results you have to split-test them in real-time (A/B split over the same period). A basic redirect usually doesn't work - Google will probably catch that quickly.